Evolutionary Neural Network-Based Online Ecological Governance Monitoring of Industrial Water Pollution DOI Creative Commons

Ying Zhao

International Journal of Swarm Intelligence Research, Год журнала: 2025, Номер 16(1), С. 1 - 23

Опубликована: Фев. 26, 2025

This paper proposes ENNOEIGS, an evolutionary neural network-based online ecological industrial governance system that integrates advanced architectures with optimization for robust pollution monitoring. The framework combines convolutional networks dimensional reduction of sensor data, external attention mechanisms discovering pattern correlations, and long short-term memory modeling the spatiotemporal evolution contaminants. A genetic algorithm continuously optimizes network parameters, enabling adaptation to changing conditions. Experimental validation using wastewater monitoring data demonstrates ENNOEIGS's superior performance, achieving a 94.8% anomaly detection rate 2.3% false alarms, outperforming existing approaches. reduces mean modified absolute error 0.028 mg/L while maintaining faster convergence during training.

Язык: Английский

Evolutionary Neural Network-Based Online Ecological Governance Monitoring of Industrial Water Pollution DOI Creative Commons

Ying Zhao

International Journal of Swarm Intelligence Research, Год журнала: 2025, Номер 16(1), С. 1 - 23

Опубликована: Фев. 26, 2025

This paper proposes ENNOEIGS, an evolutionary neural network-based online ecological industrial governance system that integrates advanced architectures with optimization for robust pollution monitoring. The framework combines convolutional networks dimensional reduction of sensor data, external attention mechanisms discovering pattern correlations, and long short-term memory modeling the spatiotemporal evolution contaminants. A genetic algorithm continuously optimizes network parameters, enabling adaptation to changing conditions. Experimental validation using wastewater monitoring data demonstrates ENNOEIGS's superior performance, achieving a 94.8% anomaly detection rate 2.3% false alarms, outperforming existing approaches. reduces mean modified absolute error 0.028 mg/L while maintaining faster convergence during training.

Язык: Английский

Процитировано

0